Source quality
Choose the right input format and avoid destructive re-encoding.
Audio FabricStudio
A practical study hub for formats, loudness, cleanup, speech, music, metadata, and repeatable sound profiles.
Cleans background noise while keeping tone natural.
Audio finishing is a chain of decisions: source format, cleanup, loudness, dynamics, destination, metadata, and final export. These guides teach the practical version of that chain.
Choose the right input format and avoid destructive re-encoding.
Read loudness, true peak, duration, and noise problems before changing the sound.
Match the chain to podcast, music, audiobook, course, restoration, or custom needs.
Use before/after listening, metadata, and export format checks before sharing.
Each guide is written around real creator decisions instead of abstract textbook definitions.
Learn when to use uncompressed, lossless, and lossy formats, how containers differ from codecs, and why delivery format depends on the destination.
Read guideUnderstand the measurements behind perceived volume, clipping risk, and why a master should be loud enough without being crushed.
Read guideA practical order of operations for hiss, hum, rumble, room tone, mouth noise, reverb, and uneven speech.
Read guideHow to move from raw audio to a finished master using diagnosis, sound profiles, before/after listening, metadata, and final export.
Read guideHow speech-oriented audio differs by destination: podcasts, interviews, audiobooks, courses, voiceover, captions, and transcription.
Read guideHow to think about finished mixes, streaming loudness, restoration cleanup, high-resolution exports, and release metadata.
Read guideA practical guide to building repeatable profile chains for shows, clients, releases, courses, or production teams.
Read guideHow titles, artists, albums, artwork, ISRC, file names, and export formats affect whether a finished master is ready to share.
Read guideStart with the destination. A podcast, audiobook, song, classroom lecture, transcription file, and restoration job should not all use the same processing chain.
| Content | Primary issue | Start with |
|---|---|---|
| Podcast | Uneven speech and background noise | Podcast Voice or Interview Balance |
| Audiobook | Long-form listening comfort | Audiobook Focus |
| Course | Clear lessons from varied recording setups | Lecture Clean |
| Music | Release polish and controlled loudness | Music Master or Music Glue |
| Restoration | Noise and fragile source material | Studio Clean or Noise Reduction |
| Recurring production | Consistent house sound | Custom Profile |
Short definitions for the measurements and controls that matter most when comparing previews.
A loudness unit that estimates perceived volume over time.
The digital audio scale where 0 dBFS is the sample ceiling.
True peak level, estimating peaks between samples.
The distance between quiet and loud moments.
How many samples per second describe the waveform.
How much amplitude detail each sample can store.
How much data per second a compressed file uses.
The method used to encode or compress audio.
The file wrapper, such as WAV, M4A, AIFF, or OGG.
Compression that preserves the original audio data.
Compression that discards data to reduce file size.
Distortion caused when a signal exceeds available headroom.
A processor that controls peaks near the output ceiling.
A processor that reduces harsh sibilance in speech or vocals.
A filter that removes low-frequency rumble below a chosen point.
Metadata fields commonly used in MP3 files.